Nearly 20% of Medicare fee-for-service patients are rehospitalized within 30 days of discharge. However, evidence suggests that as many as three-quarters of such rehospitalizations are preventable. To prevent patients suffering through unnecessary hospital readmissions, we need better understanding of the causes of and ways to prevent these avoidable events. Geriatric patients are particularly vulnerable in the posthospitalization period having intensified requirements in physical, sensory, cognitive, social and emotional function. Improved characterization of how patient function influences the likelihood of rehospitalization can facilitate more effective interventions to reduce preventable rehospitalization. To better understand the association between patient function and clinical outcomes, the NIH has commissioned the development of comprehensive measures of patient-reported physical, cognitive, sensory, social and emotional function through the NIH Patient Reported Outcomes Measurement Information System (PROMIS). Northwestern University is the statistical coordinating center for this initiative. Our central hypothesis is that poor patien self-reported health and function at the time of discharge is associated with significantly increased risk of rehospitalization, and this risk could be ameliorated with adequate support. To test this hypothesis we will prospectively measure the strength of association between self-reported health and function and rehospitalization and compare the accuracy of models of rehospitalization risk which sequentially include biomedical variables and patient-reported outcome measures of health and function. Ten functional domains will be measured at discharge and 72 hours later using validated computer adaptive testing modules developed by the PROMIS group. Multivariate logistic regression will be used to measure the relationship between these domains and rehospitalization. The accuracy of models incorporating functional predictors of rehospitalization will be compared to established biomedical models. The impact of successful completion of this research will be improved characterization of the role of self-reported health and function in older adults immediately following hospitalization allowing for more effective: 1) readmission risk prediction, 2) risk-adjusted comparison of hospital readmission performance, and 3) development of interventions to reduce rehospitalization. Current federal risk-adjustment of rehospitalization rates for purposes of performance-based reimbursement does not account for functional status at the time of discharge or in the post-discharge period. Understanding the impact of functional status in this period is necessary for fair judgment of hospital quality as measured by risk-adjusted rehospitalization rates.